160 



S. A. m\\— Tornadoes and naiI.-<fonns of Ai.ril [No. 2, 



jueiits being oi'tcu accompanied by -wLirlwiuds of any size from the 

 little dancing " dovil " or bhdt to a big dust-storm. As long, liowever, 

 as they do not extend up to the cloud layer, they produce no precipita- 

 tion, draw upon no important source of energy, and give rise to no 

 Tiolent tornadoes. 



At 10 A. M. on the 30th, the conditions were similar to those just 

 described, though the temperature over the plain -was considerably lower 

 than on the previous day ; but by 4 p. m. a rapid decrease of tempera- 

 ture at high levels had commenced, and above 5,000 feet the conditions 

 ■were such as to produce convection currents extending upwards to an 

 unknown height. During the night the greater nocturnal fall from 

 the mean temperature at the lower stations restored the conditions of 

 stable equilibrium, but by 10 A. m. on the 1st May unstable conditions 

 again appeared at heights of 7,500 feet and upwards, and by 4 p. ii. these 

 conditions extended down to below 5,000 feet. On the 2nd May, the 

 temperature at 10 A. M. decreased uniformly upwards at the rate of 

 4"0° for 1,000 feet — a rate consistent with stable equilibrium ; but in 

 the afternoon conditions likely to produce convective disturbances reap- 

 peai-ed at heights above 6,000 feet. On the 3rd, the normal conditions 

 were completely re-established. 



It would therefore -seem that the occurrence of the tornadoes of 

 the 30th April and 1st May was intimately connected with a suddon fall 

 in the temporatm-e of the air from 5,000 or 6,000 f oet upwards — a fall so 

 rapid that powerful vertical convection currents were certain to super- 

 vene. What the cause of this fall of temperature may have been is not 

 known, but presumably it was connected with the influx of easterly 

 or south-easterly winds of a moderate tempei-ature, at a time when the 

 tempei'ature at the level of the plains was excessively high. On the 

 30th, the disturbances, which were probably due to this relatively low 

 temperature at high levels, commenced i-ather late in the afternoon — 

 after 4 p. m. ; bnt on the 1st May, when unstable conditions appeared as 

 early as 10 a. m., the tornadoes took place at an earlier hour, probably 

 commencing about 2 p. m. 



When no source of energy is drawn upon in an atmospheric distur- 

 bance, except the sensible heat of the air and its actual energy of 

 motion, the disturbance is not likely to be very violent or of long dura- 

 tion but when the indefinitely large supply of energy which is stored 

 in the form of the latent heat of vapour begins to be converted into 

 the kinetic form, the disturbance may be greatly intensified and pro- 

 longed to an indefinite extent. Hence the dust-storms of the hot 

 weather never acquire any great violence, unless they extend upwards to 

 such a height that condensation of vapour occurs, and in very dry 



